Dual-Layer Strategy
Algorithmic prediction identifies candidate off-target sites, while GUIDE-seq validates real wet-lab editing events to reduce false positives and missed low-frequency events.
Gene Editing Safety Validation
Gene Editing Off-Target Assessment
Computational prediction plus wet-lab validation - identify, locate, and characterize off-target edits before you commit to clinical development.
Core Advantages
Off-target safety evidence for gene editing programs.
Use a dual-layer strategy that combines genome-wide computational screening with GUIDE-seq wet-lab validation to support clinical safety decisions.
Precise Site Mapping
Each off-target event is mapped to genomic coordinates and assessed for proximity to key genes, regulatory regions, or safety-relevant loci.
Safety-First Reporting
Off-target sites are classified by risk level and packaged into a report that supports IND safety sections and guide RNA prioritization.
Flexible Engagement
Order as a standalone safety service or embed it into AAV genome optimization and RUO manufacturing projects carrying gene editing payloads.
Editing Safety
Off-target assessment combines wet-lab validation with reviewable sequencing evidence.
Guide candidates, validation samples, sequencing outputs, and final risk summaries stay connected for therapeutic safety decisions.
Guide reviewWet-lab validationSequencing evidence
Methods & Specifications
Prediction narrows the search; GUIDE-seq confirms real events.
Dual-Layer Validation Strategy
Designed to identify potential off-target sites, verify true editing events, and classify functional risk.
| Layer | Method | Role | Description |
|---|---|---|---|
| Layer 1: Algorithmic Prediction | Proprietary prediction plus public-tool cross-checks | Identify candidate off-target sites | Uses sgRNA and genome sequence homology to produce a candidate site list with PAM, mismatch count, and position information. |
| Layer 2: Wet-Lab Validation | GUIDE-seq | Validate true off-target events | Captures genome repair events after nuclease cutting through high-throughput sequencing and maps experimentally verified off-target sites. |
Algorithmic Prediction
- Genome-wide off-target site scan based on sgRNA and reference genome sequence alignment.
- Risk scoring with mismatch count, mismatch position, chromatin accessibility, and genomic context.
- Genomic annotation for genic, non-genic, regulatory, and repetitive regions.
- Cross-reference against known functional elements and disease-associated genes.
GUIDE-seq Validation
- Captures true off-target cleavage events with GUIDE-seq labeling and sequencing.
- Quantifies off-target frequency using sequencing read counts.
- Identifies edit type, including insertion, deletion, and substitution profiles.
- Maps sites to supported reference genomes such as GRCh38 or mm39.
Sample Requirements
Matched controls help remove background signal and support clear interpretation.
| Sample Type | Description | Requirement |
|---|---|---|
| sgRNA/Cas expression system | Customer gene editing system. | Provide sgRNA sequence, Cas variant information, and expression vector sequence. |
| Edited cell system | Cells treated with the gene editing system. | Provide cell type, editing conditions, and timepoint. |
| Control cells | Untreated matched control cells. | Used for background noise filtering. |
Confidentiality note: specific algorithm models, GUIDE-seq experimental details, and bioinformatics workflow details are not externally disclosed to protect methodology IP.
Workflow
Six steps from system review to safety consultation.
Customers provide sgRNA sequence, Cas variant, and cell system information. Crossilico confirms feasibility and study design.
Standard timeline: 4-6 weeks after project confirmation. If GUIDE-seq libraries or sequencing data already exist, the bioinformatics-only workflow can shorten the timeline to 2-3 weeks.
Deliverables
Off-target safety report plus raw and structured data.
DeliverableFormatDescription
Off-Target Safety Assessment ReportPDF
Predicted site list, GUIDE-seq validated results, risk classification, functional impact assessment, method notes, and limitations.
Predicted Off-Target Site ListCSV / Excel
Candidate site coordinates, mismatch count, PAM sequence, risk score, and genomic annotation.
Validated Off-Target Site ListCSV / Excel
GUIDE-seq verified sites with coordinates, editing frequency, edit type, and genomic annotation.
Raw Sequencing DataBAM / CRAM
Raw alignment files for visualization and independent review.
Visualization FiguresPNG / PDF
Genome distribution maps, risk-level heatmaps, and editing frequency plots.
Structured DataJSON / CSV
Metadata-tagged data for computational pipelines or LIMS import.
Report template: the report structure is designed to align with common FDA and EMA IND expectations for gene editing safety data.
Use Cases
Safety evidence for IND filings, guide selection, and independent validation.
Preclinical Safety Assessment for IND Filing
Generate regulatory-ready off-target evidence for CRISPR/Cas9 and other gene editing systems.
Guide RNA Screening & Prioritization
Compare candidate sgRNAs by off-target risk profile and choose safer guides before development investment.
Add-On to AAV-Mediated Gene Editing Projects
Integrate gene editing safety data into AAV vector optimization or RUO project reports.
Independent Safety Validation
Use Crossilico as an independent third-party lab for objective off-target risk assessment.
Build Service Integration
Embed off-target assessment into AAV projects or purchase independently.
Embedded in AAV Genome Optimization & Validation
- Add off-target assessment when an AAV vector carries CRISPR/Cas or another editing payload.
- Start safety analysis during the data assessment phase.
- Integrate vector optimization data and gene editing safety data into one project report.
- Advantage: synchronized timeline, consistent data, and unified reporting.
Embedded in AAV RUO Manufacturing
- Complete off-target validation before batch release when RUO batches are used for gene editing applications.
- Attach gene editing safety data to each batch when required.
- Useful for multi-batch production programs requiring long-term safety monitoring.
Standalone Purchase
- Submit an existing gene editing system from any source for independent off-target assessment.
- Suitable for third-party audit, supplier evaluation, pipeline review, and regulatory communication support.
- Advantage: independent, objective, and fast to initiate.
FAQ
Common questions.
What is off-target editing, and why does it matter?
Off-target editing occurs when a gene editing system cuts DNA at unintended genomic locations. It can create unwanted mutations, rearrangements, or disruption of critical genes, so regulators expect comprehensive safety data before clinical trials.
How is this different from computational prediction alone?
Prediction tools are useful for screening, but they can create false positives and miss context-dependent events. Our dual-layer strategy combines prediction with GUIDE-seq wet-lab validation to report experimentally confirmed events.
What gene editing systems do you support?
We support CRISPR/Cas9, high-fidelity Cas9 variants, Cas12a, base editors, and prime editors. For non-standard systems, contact us for feasibility review.
Can you assess off-target editing in animal models?
Yes. We can analyze samples from species with available reference genomes, including mouse, rat, and non-human primates. In vivo designs require project consultation to define sample collection and depth requirements.
How does this relate to AAV Integration Risk Assessment?
Off-Target Assessment evaluates unintended DNA cuts by the editing system. Integration Risk Assessment evaluates whether the AAV vector integrates into the host genome. For AAV-delivered gene editing systems, both risks are often assessed together.
Can I add this to an existing AAV project?
Yes. If your AAV project involves a gene editing payload, add Off-Target Assessment before project completion so safety data can be integrated into the final project report.
What is the detection limit?
Sensitivity depends on sequencing depth, cell type, and editing efficiency. Detection limits are assessed during project consultation and stated in the final report.
How long does the analysis take?
Standard turnaround is 4-6 weeks from project confirmation. Bioinformatics-only analysis from existing GUIDE-seq data typically takes 2-3 weeks.
What data formats do you deliver?
Standard delivery includes a PDF report, CSV site lists, BAM/CRAM raw sequencing files, and PNG/PDF visualizations. Custom JSON with metadata is available on request.
Is the report suitable for FDA/EMA IND submission?
Yes. The strategy and report structure align with common gene therapy IND expectations for gene editing safety, including predicted and experimentally validated sites, frequency data, and risk assessment.
Start Assessment
Need gene editing off-target assessment?
Submit your sgRNA and system details for a free feasibility and project consultation. Standalone or bundled with your AAV project - flexible to your needs.